Many eukaryotic cell functions, including signal transduction, cell adhesion, gene transcription, RNA splicing, apoptosis and cell proliferation, are controlled by protein phosphorylation. Protein phosphorylation is in turn regulated by the dynamic relationship between kinases and phosphatases. Considerable research in synthetic chemistry has focused on protein kinases. However, recent biological evidence for multiple regulatory functions of protein phosphatases has triggered further investigation of phosphatases. The protein phosphatases represent unique and attractive targets for small-molecule inhibition and pharmacological intervention.
Most eukaryotic amino acid phosphate derivatives in polypeptides and proteins are found on serine, threonine and tyrosine residues. Three basic types of eukaryotic protein phosphatases have been defined: serine/threonine protein phosphatases (PSTPases), tyrosine protein phosphatases (PTPases), and dual-specificity phosphatases (DSPases). The DSPases dephosphorylate tyrosine and threonine residues on the same polypeptide substrate.
The serine/threonine protein phosphatases (PSTP ases) are further classified into subfamilies (PP1, PP2A, PP2B, PP2C and PP3) by substrate specificity, metal ion dependence and sensitivity to inhibition. At least forty different enzymes of this type have been identified through DNA cloning. Potent inhibitors of the serine/threonine phosphatases have been identified, including proteins designated Inhibitor-1, Inhibitor-2, DARPP-32, and NIPP-1, which are reviewed for example by Honkanen et al. in Protein Kinase C, Kuo, ed., Oxford Univ. Press, Oxford, 1994, p.305. In addition, several toxins, mostly from marine organisms, have been identified as potent inhibitors of the serine/threonine phosphatases. The natural product inhibitors are depicted in FIG. 1 and discussed for example by Fujiki et al. (1993) Gazz. Chim. Ital. 123: 309.
Okadaic acid, a polyether fatty acid produced by several species of marine dinoflagellates, reversibly inhibits the catalytic subunits of serine/threonine phosphatase subtypes PP1, PP2A and PP3. However, okadaic acid does not rapidly penetrate cell membranes and accumulates within cells slowly, making it difficult to control the intracellular concentration of the compound. Further, okadaic acid is not very chemically stable.
Other natural product inhibitors have been identified that are more stable, may penetrate some cell types better, are more potent, and exhibit selectivity toward different PSTPase isotypes. Calyculin A is a cytotoxic component of the marine sponge Discodermia calyx. It has an extremely high affinity to PP1, PP2A and PP3, with an inhibitory concentration.sub.50 (IC.sub.50, the concentration that causes 50% inhibition compared to untreated control preparation) being about 0.3 nM. Microcystins are potent cyclic hepta- and pentapeptide toxins of the general structure cyclo[D-Ala-X-D-erythro-b-methyl-iso-Asp-Y-Adda-D-iso-Glu-N-methyldehydro- Ala] wherein X and Y are variable L-amino acids. Microcystins are known to promote tumors in vivo, but, with the exception of hepatocytes, are impermeable to most cells in vitro. Compounds of the nodularin series exhibit IC.sub.50 's for PP1 and PP3 of about 2 and 1 nM, respectively. Motuporin, which has been recently isolated from a New Guinea sponge, is even more potent, with an IC.sub.50 of less than 1 nM for PP1. Tautomycin is produced by a terrestrial Streptomyces strain, and inhibits PPI, PP2A and PP3 indiscriminately with an IC.sub.50 in the 15 nM range. The remaining natural product inhibitors, thyrsiferal-23-acetate and cantharidine, are somewhat selective, but weak (IC.sub.50 of 0.16-10 .mu.M), inhibitors of PP2A.
High toxicity, especially hepatotoxicity, is commonly found with the naturally occurring serine/threonine phosphatase inhibitors. The high toxicity appears to be intrinsically associated with non-specific phosphatase activity, and often limits the range of feasible pharmacological studies. Honkanen (1994) Toxicon 32: 339. Further, the chemical diversity of compounds obtained from natural sources is limited. Accordingly, there is a need in the art to diversify the chemical complexity of the natural products and to optimize biochemical and pharmacological effects.
However, only limited structure-activity relationship (SAR) studies have been reported on naturally occurring serine/threonine phosphatase inhibitors. For example, SAR studies of okadaic acid indicate that the carboxyl group as well as the four hydroxyl groups are important for activity. Nishiwki et al. (1990) Carcinogenesis 11:1837; Takai et al. (1992) Biochem J. 284:539; Sasaki et al. (1994) Biochem J. 288:259.
A limited SAR study of naturally occurring microcystins was performed by Rinehart et al. (1994) J. Appl. Phycol. 6: 159. It was found that the substitution of alanine for arginine has little effect on phosphatase inhibitory potency, but does result in a difference in relative cytotoxicity. The dehydroamino acid residue and the N-methyl substituents were also found to be noncritical. Esterification of the glutamic acid residue led to inactive compounds, and the (6Z) Adda isomer was inactive, suggesting the criticality of the glutamic acid unit and the overall shape of the Adda residue. However, some variations in the Adda unit, for example the O-demethyl and the O-demethyl-O-acetyl analogs, exerted little effect on bioactivity. The general SAR of the nodularin series appears similar to the microcystins, although fewer compounds are available for testing. SAR studies have not been reported to date for calyculin A, tautomycin or thyrsiferyl acetate.
The DSPase class of phosphatases has recently been defined, and its members are emerging as important regulators of cell cycle control and signal transduction. The first documented DSPase, VH1, as described by Guan et al. (1992) Proc. Natl. Acad. Sci. USA 89: 12175, corresponds to the H1 open reading frame of Vaccinia virus. Other members of the DSPase class have been identified and generally fall into two substrate motifs, the VH1 type and the CDC25 type. Mammalian cells contain at least three cdc25 homologues (cdc25A, cdc25B and cdc25C). The CDC25 phosphatases are positive regulators of cell cycle progression, and are reviewed by Hunter et al. (1994) Cell:573. Further, there is a strong link between overexpression of the CDC25 phosphatases and oncogenic transformations, particularly in human breast cancer. Galaktinov et al. (1995) Science 269: 1575. However, no potent inhibitors of the DSPases are known.
Since nearly all forms of human neoplasias have altered cell cycle control, the role of phosphatases in cell cycle control makes these molecules attractive targets for pharmaceutical intervention. The ability of phosphatase inhibitors to interfere with aberrant cell activity has been demonstrated. For example, the naturally occurring PSTPase inhibitor okadaic acid has been shown to induce apoptosis in myeloid leukemia cells (Ishida et al. (1992) J. Cell. Physiol. 150: 484) and in rat hepatocytes, rat pituitary adenoma cell, human mammary carcinoma cells and human neuroblastoma cells (Boe et al. (1991) Exp. Cell Res. 195: 237). Thus there is a significant need to design and synthesize selective modulators of this family of enzymes in order to identify useful therapeutic agents.